Text archives Help
- From: Michael Callahan <callahan@sci.utah.edu>
- To: Thomas Schultz <schultz@mpi-sb.mpg.de>
- Cc: scirun-users@sci.utah.edu
- Subject: Re: [SCIRUN-USERS] A number of bugs in SCIRun/BioTensor fixed
- Date: Thu, 06 Oct 2005 11:49:01 -0600
Thank you Thomas. I'll integrate these changes back into the main
SCIRun trunk. Let us know if you find any other problems.
Michael
On Thu, 2005-10-06 at 15:47 +0200, Thomas Schultz wrote:
>
Hi!
>
>
I am using the BioTensor PowerApp for my master thesis (on DT-MRI
>
imaging) at Saarland University / Max-Planck-Institut Informatik. I found
>
it very useful so far; however, I came across a couple of bugs during my
>
work and I would like to send you a patch that fixes them (hope it's okay
>
to send it to the list, it's not very large).
>
>
Short description for SCIRun-bugs.diff:
>
>
1. Parts of the network would occasionally block (wait for input)
>
forever. I typically got this after turning off one of the planes and
>
turning it on again. I found this problem is due to a race condition
>
in the thread implementation: If a module sets its "need_execute" flag
>
while the scheduler runs, this can happen after the dependency checks,
>
but before the actual execution. In this case, a module gets executed
>
without its dependencies being fulfilled. I fixed this by introducing
>
a mutex within the scheduler which a module needs to hold while
>
changing its "need_execute" flag.
>
>
2. The ClipByFunction module does a test to determine if it really needs
>
to re-execute. This test should depend not only on a change in the
>
input field and clipping function, but also on the user-defined
>
variables u0-u5.
>
>
Short description for BioTensor.diff:
>
>
1. A typo in the configuration of m137 caused an error when reading in
>
tensor fields.
>
>
2. When clipping by planes, I think turning off planes altogether should
>
have the same effect as turning off each individual plane. (I exploit
>
the new behaviour when getting a glyph representation of the whole
>
volume without having to turn off each plane separately.)
>
>
3. If the isovalue changes, the clipping function (for glyphs and
>
fibers) needs to be re-evaluated because it depends on it.
>
>
4. In case the isovalue changes, the isosurface is not shown, but the
>
planes are clipped to it, the isosurface still needs to be
>
re-calculated to get the clipping right.
>
>
Thank you for a useful application. I hope my patch can contribute to its
>
further improvement.
>
>
Kind regards
>
Thomas Schultz
===========================================================================
== The SCIRun Users mailing list: send email to majordomo@sci.utah.edu ==
== for more details. ==
== Please acknowledge use of SCIRun in your papers and reports: ==
== see
http://software.sci.utah.edu/scirun-biopse_citation.bib ==
===========================================================================
Archive powered by MHonArc 2.6.16.